Skip to main content
. 2017 Nov 2;7:110. doi: 10.1186/s13613-017-0332-z

Table 2.

The baseline characteristics of the patients after propensity score matching (n = 258)

Factors Monotherapy group
n = 129
Combination therapy group
n = 129
P value Effect size
Age, mean (SD) 73.3 (12.0) 73.8 (11.5) 0.723 0.044a
Sex (n, %)
 Female 54 (41.9) 52 (40.3) 0.899 0.064a
 Male 75 (58.1) 77 (59.7)
Infection focus (n, %)
 Respiratory system 58 (45.0) 38 (29.5) 0.014 0.160
 Gastrointestinal system 22 (17.1) 34 (26.4) 0.096 0.113
 Biliary system 19 (14.7) 15 (11.6) 0.581 0.046
 Urinary system 12 (9.3) 23 (17.8) 0.068 0.125
 Musculoskeletal 8 (6.2) 4 (3.1) 0.376 0.074
 Skin and soft tissue 5 (3.9) 3 (2.3) 0.722 0.045
 Central nerve system 0 1 (0.8) 1.000 0.062
 Others 3 (2.3) 8 (6.2) 0.216 0.096
 Unknown 14 (10.9) 12 (9.3) 0.837 0.026
Surgical intervention 8 (6.2) 19 (14.7) 0.040 0.139
Non-surgical drainage# 1 (0.8) 4 (3.1) 0.370 0.084
Baseline SOFA score median [25, 75%]
 Total SOFA 10.7 [2.0, 22.0] 10.8 [3.0, 19.0] 0.849a 0.025a
 Hepatic 1.0 [0.0, 3.0] 0.0 [0.0, 3.0] 0.015 0.182
 Cardiovascular 3.0 [0.0, 4.0] 3.0 [0.0, 4.0] 0.266 0.151
 CNS system 2.0 [0.0, 4.0] 2.0 [0.0, 4.0] 0.678 0.026
Renal system 1.0 [0.0, 4.0] 1.0 [0.0, 4.0] 0.656 0.028
 Respiratory system 2.0 [0.0, 4.0] 2.0 [0.0, 4.0] 0.266 0.069
Baseline DIC score Median [25, 75%]
 Total DIC score 5.7 [2.0, 8.0] 5.7 [2.0, 8.0] 0.935a 0.013a
 SIRS score (n, %)
  0 point 51 (39.5) 37 (28.7)
  1 point 78 (60.5) 92 (71.3) 0.088 0.114
 Platelet score 3.0 [0.0, 3.0] 3.0 [0.0, 3.0] 0.778 0.048
 FDP score 3.0 [0.0, 3.0] 3.0 [0.0, 3.0] 0.829 0.105
 PT ratio score (n, %)
  0 point 15 (11.6) 21 (16.3)
  1 point 114 (88.4) 108 (83.7) 0.369 0.067
Baseline laboratory score mean (SD)
 Platelet count (× 109/L) 8.60 (6.29) 8.47 (5.04) 0.853 0.023
 FDP (μg/mL) 73.9 (191.1) 51.11 (73.35) 0.206 0.158
 D-dimer (μg/mL) 28.3 (48.9) 21.1 (24.3) 0.166 0.185
 PT ratio 2.42 (8.68) 1.50 (0.41) 0.232 0.150
 Antithrombin activity (%) 45.6 (14.1) 47.7 (11.7) 0.928 0.011a

# For the selection of covariates, a variance inflation factor (VIF) was calculated and the covariates with VIF ≧ 5 were excluded. Finally, age, sex, baseline SOFA score, baseline DIC score, and antithrombin activityused were selected.

aConfounding covariates used by the propensity score (age, sex, baseline SOFA score, baseline DIC score, and antithrombin activity)

Non-surgical interventions are as follows: percutaneous transcatheter abscess drainage, urinary tract stenting, biliary tract stenting

When the basic assumptions of Student’s t test were satisfied, data were shown mean (standard deviation) and the effect size was calculated using Cohen’s d. When the basic assumptions of Student’s t test were not satisfied, Mann–Whitney U test was performed and data were shown median [25 percentiles, 75 percentiles]. And the effect size was calculated using the following formula, Z-scores/a square root of sample number. For two-by-two contingency table, phi coefficient was used

n number, SD standard deviation, SOFA sequential organ failure assessment, DIC disseminated intravascular coagulation